Like Mike Bernstein, we have experienced a certain annoyance at
the phase shift induced by the "offset" feature of shaped pulses
generated within UXNMR (AMX). Mike wrote:
> Anyone who's spent some time using soft pulses on an AMX
>spectrometer (without linear amps) will know that some considerable time is
>spent in calibrating the soft pulses in terms of strength and, more
>annoyingly, w.r.t. PHASE. As you know, as you cross certain attenuation
>thresholds the frequency path changes, and the phase "error" changes.
> One very nice feature of soft pulses (tp07 menu in eda) is the
>ability to apply an offset, and have the soft pulse applied at a position
>away from the carrier. This is nice, but it, too, introduces an additional
>frequency-
>dependent phase error for any given power level. So, my question at last!
>Does anyone know of a method of calculating the phase error induced by the
>"offset freq." component to a soft pulse?
As I understand it : The frequency offset here is induced by applying
a linear phase ramp for the duration of the soft pulse, where the offset
is given by dP/dt, ie the rate of change of phase.
So, for an offset, dW, and a pulse of duration Pt (sec), the total phase
change is dW*Pt*360 degrees.
In principle, a major advantage of the use of phase ramps for generating
frequency offset pulses, is that overall phase coherence can be
maintained, independent of the offset. This avoids the necessity for
expensive phase-locked or phase coherent synthesisers, and the need to
apply an offset-dependent phase correction to each soft pulse. For the
overall pulse phase to remain constant, the ramp must be applied
symmetrically. If the phase ramp starts at 0, the final phase value
will obviously vary according to the offset and the overall phase will
vary also. This seems to be how the method is implemented in UXNMR.
However, if the ramp is applied such that there is a 0 phase cross-
over point coinciding with the midpoint of the pulse, then the
overall pulse phase will be constant, regardless of offset. (The
Bruker Medizintechnik "pipeline" acquisition software for the
Biospec systems functions correctly in this respect!)
Any information from Bruker regarding the operation of phase
ramps, and the possibility of changes to its current implementation
to allow offset-independence, would be very much appreciated.
Regards
Ian
--
==================================================================
Ian Brereton
Centre for Magnetic Resonance
University of Queensland
St. Lucia Qld 4072
Australia
Phone: +61-7-365-4100 Fax: +61-7-365-3833
IBrereton@cmr.uq.oz.au
==================================================================